Doxorubicin, cisplatin and etoposide were purchased from Sigma-Aldrich Co. ABT-737 was purchased from Santa Cruz Biotechnology, Inc.

The neurofibromin-deficient MPNST cell line, sNF02.2, and benign the neurofibroma cell line, Hs 53.T, were purchased from the American Type Culture Collection and grown in DMEM media (Hyclone Laboratories) supplemented with 10% FBS (Hyclone Laboratories), penicillin (100 U/ml), and streptomycin (100 μg/ml). All cultured cells were incubated at 37°C in a humidified atmosphere containing 5% CO₂.

Normal tissue and tumor tissue specimens were obtained by skin biopsy and surgical resection, respectively, from a 24-year-old male patient with NF1 (23,24). The patient having a NF1 nonsense mutation Y2264X (c.6792C>G) in the NF1 gene presented the clinical features of NF1 with CAL spots, scoliosis, cutaneous neurofibromas, subcutaneous neurofibromas, PNs and MPNSTs (23,24). After surgical resection at 24 years of age, the patient died at the age of 25. Three types of tissues, normal phenotypic tissues, PNs and MPNSTs, were pathologically evaluated by routine light microscopy after staining with hematoxylin and eosin (H&E) as previously described (24). The study was approved by the Institutional Review Board Committee of the Ajou University School of Medicine.

Primary tissue culture was performed by primary explant technique. The dissected tissues were finely chopped, rinsed with PBS, and the pieces were seeded onto the surface of a tissue culture flask in 1 ml of DMEM supplemented with a high concentration (40%) of FBS. After an overnight incubation at 37°C, the medium volume was made up to 5 ml and then changed weekly until a substantial outgrowth of cells was observed. Cells were then grown in DMEM media supplemented with 15% FBS. Cells were used from passages 5 through 10. The primary cells from the three types of tissues, normal phenotypic tissues, PNs and MPNSTs, demonstrated their distinct cellular characteristics by western blotting with antibodies against GTP-Ras and its downstream effectors (Fig. 1).

The cell viability assay was performed by using the EZ-Cytox Cell Viability Assay kit (Daeil Lab Service, Korea). The cultured cells were plated at a density of 4×10³ in a 96-well flat-bottom tissue-culture plate, incubated overnight, and were treated with the indicated concentrations of drugs. After 24 h incubation, 10 μl of Ez-Cytox reagent was added, and the cells were incubated for another 2 h and then absorbance was measured at a wavelength of 450 nm with an ELISA microplate reader (Model 680; Bio-Rad).

The target sequences for the small interfering RNAs (siRNAs) (Genolution, Korea) were as follows: 5′-CAG TGAACGTAAGGGTTCT-3′ for the NF1 gene, 5′-CAGGGACAGCATATCAGAG-3′ for the BCL2L1 (Bcl-xL) gene and 5′-CCTACGCCACCAATTTCGT-3′ for the nonspecific negative control. The siRNAs were diluted in serum-free Opti-MEM (Invitrogen) and transfected into cells using Lipofectamine™ RNAiMax (Invitrogen).

Total-RNAs were isolated from cultured cells using TRIzol reagent (Invitrogen), treated with RNase-free DNase I (Invitrogen) to avoid amplification of genomic DNA, and were subsequently reverse transcribed by the RevertAid™ H Minus First Strand cDNA Synthesis kit (Fermentas) with oligo(dt)_15–18 primer. Real-time RT-PCR was performed using the SYBR-Green I qPCR kit (Takara, Japan). The specific primers used were as follows: 5′-GTCGGATCG CAGCTTGGATGGCCAC-3′ and 5′-CGTCAGGAACCAG CGGTTGAAGCGT-3′ for BCL2L1, P238284 primer set (Bioneer, Korea) for NF1, and 5′-TGTTGCCATCAATGA CCCCTT-3′ and 5′-CTCCACGACGTACTCAGCG-3′ for the GAPDH gene (a relative quantification standard). All real-time RT-PCR measurements were performed using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems).

Cultured cells were lysed in RIPA buffer (150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, and 50 mM Tris buffer, pH 8.0). Proteins were heated at 95°C for 5 min and analyzed by SDS-PAGE on 8–12% polyacrylamide gels. The proteins were electroblotted onto a PVDF membrane (Millipore). The membrane blots were blocked with 5% (w/v) nonfat dried milk, incubated with primary and secondary antibodies, and then were visualized by the ECL western blotting detection system (WEST-ZOL plus; Intron Biotechnology, Korea). The following antibodies were used: anti-Bcl-xL, anti-caspase-3 anti-Bcl-2, anti-Bax, anti-phosphorylated Akt, anti-Erk1/2, anti-phosphorylated Erk1/2, and anti-Ras antibodies (Cell Signaling Technology); anti-p120RasGAP (BD Transduction Laboratories); anti-neurofibromin, anti-actin, anti-p53, anti-Mcl-1, anti-K-Ras, anti-H-Ras, anti-SOCS3, HRP-conjugated goat anti-rabbit IgG and HRP-conjugated goat anti-mouse IgG antibodies (Santa Cruz Biotechnology, Inc.).

Ras-GTP was detected by using a Ras-activation assay kit (Upstate Biotechnology). Briefly, active Ras was precipitated by a GST fusion protein containing the Ras-binding domain of Raf (GST-Raf-RBD). Cells were lysed in lysis buffer (25 mM HEPES pH 7.5, 150 mM NaCl, 1% Igepal CA-630, 10 mM MgCl₂, 1 mM EDTA, and 2% glycerol). A total of 300 μg cellular lysate was incubated with Raf-1 RBD agarose at 4°C for 1 h. Agarose beads were washed three times with 1 ml of ice-cold lysis buffer, boiled with a 2X Laemmli sample buffer, and separated on SDS-PAGE gels, followed by western blot analysis using an anti-Ras antibody.

Results are expressed as the mean ± SD. All experiments were repeated at least three times. Statistical significance between the groups was calculated by a Student’s t-test. Probability values <0.05 (P<0.05) were considered statistically significant.

Understanding the mechanism of drug resistance is crucial for developing new strategies for targeted chemotherapy. To examine whether the chemosensitivity to anticancer drugs between the benign neurofibroma and NF1-associated malignant MPNST cells was different, we firstly investigated the cytotoxic sensitivity to the representative anticancer drugs inducing apoptosis in the established cell lines, Hs 53.T and sNF02.2 (25). Since doxorubicin, cisplatin and etoposide have been studied in the patients with NF1-associated MPNSTs (13–16), we used these three anticancer drugs in this study. The cell viability and caspase-3 cleavage assay results showed that the MPNST sNF02.2 cells were more resistant to all three drugs than benign Hs 53.T cells (Fig. 2A–C).

Next, we tried to confirm this result in the NF1-associated primary cells. We performed primary tissue culture of the three types of pathologically evaluated tissues; normal phenotypic tissues (PC-N), benign PNs (PC-B) and malignant MPNSTs (PC-M), derived from a patient with NF1, and demonstrated their distinct cellular characteristics by comparing the levels of GTP-Ras and its downstream effectors, phosphorylated Erk1/2 and phosphorylated Akt, and by the expression level of SOCS3, suppressor of cytokine signaling (Fig. 1). Due to poor proliferation in the primary benign PN cells after 5 passages, the primary normal cells and MPNSTs cells were used in this study. As observed in the cell lines, the primary neurofibromin-deficient MPNST cells were more resistant to all three drugs than the primary neurofibromin-deficient normal phenotypic cells (Fig. 2D–F). In order to understand the reason for the difference in drug resistance between the MPNST cells and benign/normal cells, we further investigated the expression levels of apoptosis-related proteins in these cells. Interestingly, we found that the basal expression level of Bcl-xL was significantly increased in both the MPNST cell line and primary MPNST cells compared to the benign cell line and primary normal cells, while none of the expression levels of other anti-apoptotic proteins Bcl-2 and Mcl-l were found to be different (Fig. 3A and B) in the MPNST cells compared to that in benign/normal cells.

Activated cell survival signaling linked with apoptosis prevention in the NF1-associated MPNSTs is implicated in the activation of the Ras-signaling pathway (4,6). Since the reduced expression of neurofibromin, a negative regulator of this signaling pathway, in the MPNSTs has been reported (26), we firstly investigated the basal expression levels of neurofibromin and observed that the neurofibromin expression levels were significantly lower in both the MPNST cell line and primary MPNST cells compored to the benign/normal cells (Fig. 3A and B). Quantitative RT-PCR revealed that the different expression patterns of neurofibromin and Bcl-xL proteins between the MPNST cells and benign/normal cells originated from the difference in the transcriptional expression of the NF1 and BCL2L1 genes (Fig. 3C and D). To elucidate whether the decreased expression level of neurofibromin in the MPNST cells was caused by an additional mutation in the intact NF1 allele, we performed molecular analysis of the NF1 gene in the primary MPNST cells. The presence of the normal NF1 allele besides the mutated NF1 allele in trans-chromosomes was confirmed in the primary MPNST cells (data not shown).

In order to determine if the expression level of Bcl-xL were dependent on the expression levels of the NF1 gene, we performed the NF1 gene silencing experiment in the benign/normal cells. Depletion of neurofibromin expression by siRNA treatment for the NF1 gene caused an increase in the Bcl-xL expression in a dose-dependent manner in both the benign and normal cells, but it did not have any effect on other anti-apoptotic proteins, Bcl-2 and Mcl-1. Quantitative RT-PCR analysis demonstrated that the knockdown of NF1 by RNAi led to the increase in the mRNA level of the BCL2L1 gene (Fig. 4A and C). We next examined whether the depletion of neurofibromin in the benign/normal cells had an influence on the resistance to anticancer drugs. When doxorubicin was co-treated for 24 after 72 h of NF1 siRNA treatment, the neurofibromin-depleted cells showed increased cell viability and decreased caspase-3 cleavage activity in both the benign and normal cells compared to control RNAi cells (Fig. 4B and D).

Our results demonstrated that resistance of the MPNST cells to anticancer drugs was caused by increased Bcl-xL expression. Therefore, our study focused on the manipulation of the Bcl-xL expression and the siRNAs targeted against downregulation of Bcl-xL in the MPNST cells. The BCL2L1 siRNAs per se did not have any effect on cell toxicity in both the MPNST cell line and primary MPNST cells (data not shown). However, co-treatment for 24 h with doxorubicin after 72 h of BCL2L1 siRNA treatment dramatically reduced the cell viability and increased the caspase-3 cleavage activity in both the MPNST cell line and primary MPNST cells. There was no effect when the negative control siRNAs were co-treated with doxorubicin (Fig. 5). These results indicated that the downregulation of Bcl-xL by RNAi enhanced chemosensitivity of the MPNST cells to the anticancer drug doxorubicin.

We next investigated whether ABT-737 (27), a recently developed Bcl-2 family protein specific inhibitor, had an inhibitory effect against Bcl-xL in the MPNST cells, and whether ABT-737 displayed a synergistic cytotoxicity with anticancer chemotherapeutic agents as in the co-treatment of BCL2L1 siRNA. The concentration of doxorubicin (0.5–5 μg/ml) required for effective apoptosis was estimated from the results of Figs. 2 and 5. ABT-737 alone did not have an effect on cell toxicity in both the MPNST cell line and primary MPNST cells at the concentrations tested (1–80 μM) (data not shown). However, when the cells were co-treated with doxorubicin, ABT-737 effectively enhanced apoptotic cell death in a dose-dependent manner in both the MPNST cell line and primary MPNST cells, compared to cells with single treatment of doxorubicin (Fig. 6). Notably, ABT-737 in combination with doxorubicin dramatically enhanced chemotherapy sensitivity in the MPNST cells. Furthermore, ABT-737 effectively reduced the dosage of doxorubicin required for efficacious MPNST cell death. The concentrations of both ABT-737 and doxorubicin required for causing approximately 80% cytotoxicity in the established and primary tissue cultured NF1-associated MPNST cells were calculated as; 1 μM of ABT-737 plus 2.5–5 μg/ml of doxorubicin.